Although widely used as a screening test for the early detection of colorectal cancer, the efficacy of fecal occult blood testing (FOBT), to prevent death from colorectal cancer is unknown. Randomized controlled trials have been underway for some time, but it is doubtful if an answer about FOBT will be available any time soon because of the need for larger than planned sample size, extended follow-up, problems with screening compliance, and the presence of screening among the controls. A population-based case-control study is being conducted to evaluate the efficacy of FOBT. Case (n=239) are all deaths from colorectal cancer occuring during the years 1986-1991 among Group Health Cooperative (GHC) members; controls (n=478) are randomly sampled from the GHC membership list and matched to cases according to age, gender and duration of membership. Medical chart review is used to ascertain a history of FOBT. Analysis will consist of a comparison of a history of screening in defined intervals up to the diagnosis date of the case to identify the interval during which screening is most likely to be efficacious. Analytic methods appropriate for the interpretation of screening efficacy in case-control studies will be compared and refined, if necessary. Finally, results from this study will be combined with two recently completed case-control studies to increase the precision with which the efficacy, if any, of FOBT can be estimated. These results may provide the only available evidence of FOBT efficacy for some time to come.